Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

389results about How to "Easy input" patented technology

Natural environment bird monitoring method based on multi-modal fusion deep learning and computer device

The invention relates to the field of bird monitoring, in particular to a natural environment bird monitoring method based on multi-modal fusion deep learning and a computer device. The scheme comprises the following steps: establishing a multi-modal bird data set; labeling the multi-modal bird data set; carrying out preprocessing on the marked data; the method comprises the following steps of: constructing a multi-modal trans-attention Transform network model; the method comprises the following steps of: training and optimizing a multi-modal trans-attention Transform network model; sound and image data in the multi-mode bird data set are subjected to conjoint analysis, a YOLOv8 target detection model is used, marked image data are trained, and rapid detection of a bird target is achieved; in combination with a DeepSORT tracking algorithm, the detected target is tracked, and the motion trail of the birds is obtained; and performing space-time synchronization on the sound source positioning information and the visual tracking result, and verifying the identity and the position of the bird. The bird monitoring device is suitable for monitoring birds in the natural environment.
Owner:SICHUAN AGRI UNIV

Multi-modal large language model for realizing fine-grained visual perception of remote sensing image

The invention provides a multi-modal large language model for realizing fine-grained visual perception of remote sensing images, which comprises a shared visual coding module which comprises two functional complementary visual encoders and is used for respectively receiving and coding images with different resolutions after down-sampling, and connecting the images according to channel dimensions to obtain an integrated multi-scale visual feature map. Meanwhile, respectively inputting the visual prompts into two visual encoders with complementary functions to obtain encoded visual prompts; the modal alignment mapping layer is used for converting the multi-scale feature map and the coded visual prompts into a language semantic space to obtain mapped vision and visual prompts; the text word segmentation device module is used for converting a text instruction into vector embedding; and the large language model decoder is used for receiving the mapping image token and the visual prompt token, embedding a text instruction and generating a model response sequence. The invention provides a simple and feasible method for research and application of a fine-grained multi-modal large language model in the field of remote sensing.
Owner:BEIJING INST OF TECH

Deep concealed deposit detection method based on multi-source geological data fusion

The invention provides an innovative deep concealed ore deposit detection method, which comprehensively utilizes geology, geophysics, geochemistry and remote sensing technologies, and realizes effective integration of multi-source geological data by developing an advanced data fusion algorithm. According to the method, data collected by geological survey, wide-area electromagnetic method exploration, hydrocarbon-mercury superposition halo analysis and aviation hyperspectral remote sensing are subjected to standardized preprocessing, so that the consistency and accuracy of the data are ensured; and then, deep analysis is carried out on the fused data by utilizing machine learning and statistical analysis technologies so as to identify and predict mineralization modes and abnormal regions. Finally, a three-dimensional geologic model is constructed, the form and distribution of an ore deposit are visually displayed, and a scientific basis is provided for geological exploration. The method is characterized in that a multi-source data fusion technology and mineralization mode identification automation and precision are realized, so that the efficiency and the accuracy of deep hidden ore deposit detection are remarkably improved.
Owner:湖南省遥感地质调查监测所

Real-time monitoring method for abnormal operation state of energy storage power generation system

The invention provides a real-time monitoring method for an abnormal operation state of an energy storage power generation system, and relates to the technical field of energy storage monitoring, and the method comprises the steps: collecting operation parameters of a battery pack through a first sensor group; acquiring environmental parameters of the battery pack through a second sensor group; uploading the key parameters and the environmental parameters to a platform; the cloud platform performs first judgment on the operation parameters through a control model of each parameter to obtain an abnormal mark; according to the operation parameters, a fault is predicted through a cloud platform and a fault prediction model, and a fault prediction result is obtained; according to the abnormal mark and / or the fault prediction result, an early warning is given, and an energy storage system adjustment suggestion is given; according to the method, operation parameters and environmental parameters of the battery pack can be comprehensively considered, fault prediction is carried out by dynamically adjusting the control range and utilizing machine learning, and specific adjustment suggestions are given, so that the safety and reliability of the energy storage system are improved.
Owner:BEIJING WARMLAND ENERGY SERVICE CO LTD

Intelligent question answering system and method for cross-domain knowledge reasoning

The invention discloses an intelligent question answering system and method for cross-domain knowledge reasoning, and relates to the technical field of intelligent question answering. In order to solve the problem that an existing intelligent question answering system is limited by technical shortages and cannot give satisfactory reply to a user in cross-domain question processing, the adopted scheme comprises the steps that a domain knowledge base construction module collects multi-domain knowledge data, and the multi-domain knowledge data is stored in a corresponding domain vector knowledge base after being preprocessed and vectorized; the user interaction module provides an interface, receives input, interacts and returns a result; the field identification module identifies the field and extracts key information by using a large language model; the knowledge retrieval module retrieves related information accordingly; the cross-domain knowledge reasoning module is used for reasoning a knowledge logic relationship of each domain and guiding a large language model to integrate fused knowledge; and the intelligent question-answering module generates prompt words, inputs the prompt words into the large language model to obtain a final response, and returns each result to the user interaction module for display. According to the method, knowledge from multiple fields can be processed, and the method is suitable for various different types of problems and scenes.
Owner:SHANDONG INSPUR SCI RES INST CO LTD

Radio frequency front-end module and radio frequency chip

The invention relates to the technical field of wireless communication, and provides a radio frequency front-end module and a radio frequency chip, the radio frequency front-end module comprises a substrate, and a first chip, a second chip and a third chip arranged on the substrate; the first chip is provided with a power amplification circuit which is used for carrying out amplification processing on a sent radio frequency signal. A first voltage conversion circuit is arranged on the second chip and is used for converting a power supply into a preset voltage and outputting the preset voltage to the first chip to provide a power supply; a control module is arranged on the third chip, and is used for processing the amplification signal output by the power amplification circuit and outputting a corresponding control signal to realize switching; the output end of the control module is used for controlling to switch different radio frequency signals to the signal receiving end; the input end of the first voltage conversion circuit is used for being connected with a power supply, and the first voltage conversion circuit is used for converting the power supply into working voltage and outputting the working voltage to the power amplification circuit. According to the radio frequency front-end module, the switching effect and reliability of the transmitted and received signals can be improved.
Owner:LANSUS TECH INC

Thermal power plant DCS production process intelligent early warning system based on artificial intelligence

The invention relates to the technical field of thermal power plant production management, in particular to a thermal power plant DCS production process intelligent early warning system based on artificial intelligence, which comprises a data acquisition module for receiving real-time production data of each production link of a thermal power plant acquired from a DCS; the data storage module is connected with the data acquisition module and is used for storing the acquired real-time production data; the data preprocessing module is used for preprocessing the stored data; the feature extraction module is used for extracting key feature parameters from the preprocessed data; the model training module is used for training an artificial intelligence model based on historical data and the extracted key feature parameters; the real-time monitoring module is used for monitoring the production process in real time; according to the invention, the method captures the dynamic change of the time sequence data and the mutual relation between the multi-dimensional features, and achieves the more accurate prediction of the future time.
Owner:HUBEI ENERGY GRP EZHOU POWER GENERATION CO LTD

Intelligent medical pre-inquiry method and system based on multi-modal large model, and medium

The invention discloses an intelligent medical pre-inquiry method and system based on a multi-modal large model and a medium, and the method comprises the steps: obtaining user input data to recognize a target intention, and when a knowledge question and answer intention is recognized, obtaining a knowledge question and answer module; based on the medical knowledge graph, combining with knowledge keywords in the user input data and utilizing a multi-modal large model to obtain medical answers; when the pre-inquiry intention is recognized, inquiry keywords in the user input data are extracted for slot filling to obtain slot information, and a slot filling strategy is dynamically adjusted based on the slot information and user feedback data; and performing visual display on the medical answers and the health management suggestions based on the user interaction interface, and obtaining user interaction data for human-computer interaction. According to the invention, a natural language processing and multi-modal fusion mode is adopted, so that the information input process of a patient is more convenient; and meanwhile, by introducing dynamic medical history information acquisition and interaction, real-time interaction with the patient is realized, and the user experience of pre-inquiry is improved.
Owner:HANGZHOU ZHIDIAN TECHNOLOGY CO LTD

Training method and reconstruction method of glaucoma fundus image super-resolution generation model

The invention discloses a training method and a reconstruction method of a glaucoma fundus image super-resolution generation model, and belongs to the technical field of medical image data processing. In order to solve the problem that when super-resolution reconstruction is carried out on a glaucoma fundus image through an existing generative model, reconstruction cannot be effectively carried out according to the characteristics of the glaucoma fundus image, the reconstruction generative model is a generator of a generative adversarial network, and the high-resolution glaucoma fundus image serves as a reference image; a low-resolution image obtained by down-sampling the fundus image is used as the input of a generator, and the generator can generate a super-resolution reconstruction image of the fundus image of glaucoma through the overall training of the generative adversarial network, and in the training process of the generative adversarial network, the super-resolution reconstruction image of the fundus image of glaucoma is obtained. And calculating image quality evaluation loss and edge similarity loss by using the cross section area of the nth nerve fiber layer of the reference image and the reconstructed image and the binary edge image of the sieve plate region as total loss to realize model training.
Owner:HARBIN INST OF TECH

Mura defect detection method and device of TFT-LCD display screen and electronic equipment

The invention discloses a Mura defect detection method and device of a TFT-LCD display screen and electronic equipment, and relates to the field of display screens. The method comprises the following steps: acquiring an original image of a TFT-LCD display screen to be detected; preprocessing the original image to obtain a target image, the preprocessing including image noise reduction and image enhancement; based on the target image, constructing a multi-scale feature map of the Mura defect, the multi-scale feature map comprising a local mean map and a local variance map under multiple scales: performing adaptive weight fusion on the local mean map and the local variance map to obtain a confidence map of the Mura defect; according to the confidence map, utilizing morphological operation to extract a suspected defect area; performing shape analysis and contrast analysis on the suspected defect area to obtain shape features and contrast features of the suspected defect area; and screening out a target Mura defect region from the suspected defect region based on the shape feature and the contrast feature. By implementing the technical scheme provided by the invention, the accuracy of Mura defect detection is improved.
Owner:ALTRON OPTOELECTRONICS (SHENZHEN) CO LTD

Steel coil edge detection machine based on machine vision

ActiveCN119044066AReduce the inconvenience of later detectionReduce the problem of cost increasePreparing sample for investigationOptically investigating flaws/contaminationMachine visionStructural engineering
The invention discloses a steel coil edge detection machine based on machine vision, which comprises an arranged moving seat, and a supporting seat is arranged at the side end of the moving seat; a storage box is installed on the supporting seat, a cleaning assembly is arranged in the storage box, and the cleaning assembly comprises a liquid storage cylinder installed in the storage box; and a driving assembly is arranged in the storage box, the special-shaped liquid bag is extruded through movement of the second moving plate, liquid in the special-shaped liquid bag is conveyed into the second pipeline through a pressure valve, a detection piece is connected to the rotating assembly, and the edge of the steel coil is detected through the detection piece. According to the steel coil edge detection machine based on machine vision, liquid is sprayed out for cleaning through the through hole in the rotating rod, the problem that later detection is inconvenient due to much dust at the side end of a transported steel coil is solved, the edge of the steel coil is detected through the detection piece, the overall structure is simple, and the problem that cost is increased due to the tedious structure is solved; and the overall practicability is improved.
Owner:ZHEJIANG YILI METAL MATERIAL CO LTD

Prediction method for intestinal preparation of patient before colonoscopy based on machine learning

PendingCN119541856AImprove clinical productivityImprove interactivityHealth-index calculationMedical automated diagnosisBowel preparationIntestino-intestinal
The invention discloses a method for predicting intestinal preparation of a patient before colonoscopy based on machine learning, and the method comprises the following steps: S1, collecting clinical data of the patient, and carrying out the preprocessing of the clinical data; s2, performing feature selection on the collected clinical data by using a Boruta algorithm, and identifying important features related to insufficient intestinal preparation before colonoscopy; s3, constructing a prediction model by utilizing a machine learning algorithm, evaluating the prediction model by adopting a cross validation method, and selecting an optimal prediction model; s4, explaining the optimal prediction model by adopting an SHAP explaining tree model, and identifying a feature which has the greatest influence on a prediction result; s5, clinical data of a to-be-detected patient are transmitted to the optimal prediction model, and intestinal preparation of the patient before colonoscopy is predicted; according to the method, the risk of insufficient intestinal preparation can be accurately predicted, personalized guidance can be provided, the clinical working efficiency is remarkably improved, and the method has important clinical significance.
Owner:ZHONGSHAN HOSPITAL AFFILIATED TO FUDAN UNIV XIAMEN HOSPITAL

Piezoelectric self-feedback excitation type miniature electric field sensor, electric field detection method and sensor

The invention provides a piezoelectric self-feedback excitation type miniature electric field sensor, an electric field detection method and a sensor, and is applied to the technical field of electric field sensors. The piezoelectric self-feedback excitation type miniature electric field sensor comprises a substrate used for supporting a vibration pickup reference structure, a piezoelectric driving structure, a mutual shielding electric field sensing structure and an elastic beam; the piezoelectric driving structure is used for enabling the piezoelectric driving structure to deform under the condition that driving voltage is applied to the piezoelectric driving structure so as to enable the elastic beam to generate vibration; the mutual shielding electric field induction structure is used for responding to the vibration of the elastic beam, and the mutual shielding electric field induction structure outputs an induction current signal in a differential form so as to suppress common-mode noise and detect the amplitude intensity of a to-be-detected electric field; and the vibration pickup reference structure is used for responding to the vibration of the elastic beam and outputting a reference current signal so as to adjust the magnitude of the driving voltage and enable the output sensitivity of the piezoelectric self-feedback excitation type miniature electric field sensor to be a preset value.
Owner:AEROSPACE INFORMATION RES INST CAS

Fruit image classification method based on deep transfer learning

The present invention relates to a fruit image classification method based on deep transfer learning, and belongs to the field of agricultural informatization. The algorithm freezes the low-level network parameters, makes the low-level network structure as a feature extractor, and modifies the high-level network parameters and structure, and proposes new migration models TL-VGG16, TL-InceptionV3 and TL-ResNet50. By inputting images into the three migration models for small sample training, accurate classification of different fruit images can be achieved. Finally, the three migration models are fused using the weighted average method and the relative majority voting method, and new prediction probabilities are output. The present invention solves the problems of traditional deep networks in image recognition, such as deep layers, long training time, slow model convergence, and serious overfitting in small data sets, and realizes lightweight parameters of deep networks, reduces the difficulty of training, and increases the universality of deep networks in smart agricultural applications.
Owner:CHONGQING UNIV OF POSTS & TELECOMM

Single-phase open-circuit fault-tolerant current reconstruction method for slice motor

The invention discloses a thin-sheet motor single-phase open circuit fault-tolerant current reconstruction method for eliminating reverse magnetic potential based on a magnetic field modulation theory. A motor structure with six teeth and one pair of poles is adopted. The method comprises the following steps: firstly, according to a calculation formula of magnetic potential, writing a magnetic potential expression of residual-phase current in an air gap after single-phase open circuit; according to the obtained magnetic potential expression, extracting a reverse magnetic potential expression in the magnetic potential expression; the expression of the reverse magnetic potential is equivalent to a magnetic vector, and the magnetic vector is divided into two groups, namely ten vectors; after the phase difference between two vectors is 180 degrees, the residual phase current counteracts each other, and the remaining three vectors form an equilateral triangle to counteract each other; the five-phase current meeting the conditions simultaneously meets the condition that the reverse magnetic potential is 0, the suspension force is kept consistent with that when fault tolerance is not carried out, and suspension force pulsation does not exist. According to the invention, the defect that the reconstruction current after the traditional single-phase open circuit is applied to the slice motor is overcome, the suspension performance of the slice motor after fault tolerance is improved, and the practicability is enhanced.
Owner:NANJING UNIV OF AERONAUTICS & ASTRONAUTICS

Conference summary generation method and device, computer equipment and storage medium

The invention relates to the technical field of data processing, and discloses a conference summary generation method and device, computer equipment and a storage medium. According to the conference summary generation method, a plurality of conference record text blocks are obtained by preprocessing a conference record text, and the conference record text blocks are input into an issue segmentation model; the method comprises the steps of obtaining an issue segmentation text, inputting the issue segmentation text into a word vector model to obtain an issue vector of each issue, finally calculating the similarity of each issue according to the issue vector of each issue, merging the issues with the similarity higher than a preset similarity, and inputting the combined issues into a language model. Therefore, the conference summary which is uniform in format, rich in content and high in value is generated, subsequent review and reference of participants are greatly facilitated, and meanwhile the team cooperation efficiency and fluency are remarkably improved.
Owner:SHENZHEN KUAILU TECH CO LTD

Environment perception data processing method and system based on intelligent helmet

The invention relates to the field of data processing, in particular to an environment perception data processing method and system based on an intelligent helmet. The method comprises the following steps: acquiring a scene image in front of a user and multi-source environment real-time sensing data; performing scene attribute state analysis on the multi-source environment real-time sensing data, and constructing a real-time scene attribute state diagram; performing contrast dynamic balance adjustment on the scene image in front of the user, performing super-resolution fusion reconstruction, and constructing a super-resolution scene image; performing object visual identification on the super-resolution scene image, and performing object bounding box segmentation processing to obtain a plurality of accurate scene object bounding boxes; and carrying out optical flow motion change analysis on the plurality of accurate scene object bounding boxes, and carrying out space-time scene structure evolution analysis so as to obtain dynamic scene evolution characteristic data. According to the method, the data processing efficiency of the environment perception data is improved, and the environment perception data is visualized more visually.
Owner:ORIGJOY

Aircraft windshield wiper rotation angle control device and control method

The invention relates to the technical field of airplane windshield wiper control, in particular to an airplane windshield wiper rotation angle control device and method, and the method comprises the steps: obtaining an initial windshield wiper vibration signal of a to-be-controlled airplane in a current windshield wiper cleaning process, decomposing the initial windshield wiper vibration signal, and screening out a target current vibration signal; determining a rotation error factor corresponding to the complete cleaning process of each windshield wiper; determining vibration abnormal factors corresponding to each current sampling moment and each historical sampling moment; determining a current error accumulation factor; according to the vibration abnormal factors corresponding to all the current sampling moments and the current error accumulation factor, the PID control quantity at the current moment is adjusted, a target PID control quantity is obtained, and turning angle control is conducted on a windshield wiper on the to-be-controlled aircraft. According to the invention, the rotation angle control of the windshield wiper is realized, and the reasonability of PID control quantity adjustment and the reasonability of rotation angle control of the wiper are improved.
Owner:XIAN QINGAN ELECTRIC CONTROL

Product recommendation method and device, electronic equipment, storage medium and program product

The invention provides a product recommendation method and device, electronic equipment, a storage medium and a program product. The method comprises the steps of obtaining a query vector; retrieving in a product vector database based on the query vector to obtain a plurality of initial product data; obtaining user basic data and historical behavior data; based on the multiple pieces of initial product data, the basic data and the historical behavior data, performing training through a product recommendation probability prediction model to obtain multiple pieces of target product data; and generating recommendation content of each piece of target product data. According to the product recommendation method provided by the invention, the query vector is generated, rapid matching with the product vector is realized in the product vector database, basic data and historical behavior data are combined on a retrieval result, model input is optimized, and meanwhile, short-term interest and long-term interest are fused in an attention mechanism of the model, so that the product recommendation efficiency is improved. Therefore, the target product data obtained based on the model can better fit the interest of the user on different time scales, and the content is recommended according to the generated accurate product.
Owner:CHINA MOBILE ONLINE SERVICES CO LTD +1

Intelligent flow regulation efficient cooling tower device

The utility model relates to the technical field of cooling towers, in particular to an intelligent flow regulation efficient cooling tower device. The device comprises a tower body, a support is fixedly connected to the position, close to the top, of the inner wall of the tower body, a wind power assembly movably communicates with the interior of the support, and multiple exhaust assemblies are movably connected into grooves in the outer wall of the tower body. The adjusting assembly rotates on the outer wall of the circulating assembly, so that different numbers of holes in the adjusting assembly are aligned with holes in the circulating assembly, liquid in the circulating assembly can conveniently penetrate through the holes in the adjusting assembly to be input into the filter barrel, and filler in the filter barrel is cooled; meanwhile, liquid discharged by the filler is recycled to the bottom of the tower body through the filtering barrel, the liquid enters the circulating assembly to be filtered and then is output again to cool the filler, the air exhaust assembly rotates in the tower body to drive air between the interior and the exterior of the tower body to circulate, and steam in the tower body is output upwards through rotation of the wind power assembly.
Owner:YICHANG ADVANCED NEW MATERIALS R&D CO LTD

Recycling and reusing device for discharged water based on RO (Reverse Osmosis) concentrated water system

The invention relates to the technical field of discharged water recycling, and discloses a discharged water recycling device based on an RO concentrated water system, which comprises a buffer tank, the buffer tank is connected with an ultrafiltration device, the ultrafiltration device is connected with a precision filter, the precision filter is connected with an RO device, the RO device is connected with a nanofiltration device, and the RO device is connected with a product water tank. The concentrated water is filtered by the RO device and then is divided into two paths of produced water, one path is recycled water, the recycled water enters a product water tank after the produced water conductivity and the produced water flow are detected, and the other path is re-concentrated water, and the re-concentrated water enters the nanofiltration device after pressure detection. Excess pressure of the concentrated water RO device is utilized to enter the nanofiltration device, nanofiltration is not provided with a booster pump, energy conservation and consumption reduction are achieved, the structure is simple and reasonable, the nanofiltration device is independent of original main system equipment, normal operation of the main equipment is not affected when the equipment breaks down or stops running, the water treatment capacity of a preceding-stage device can be reduced, waste water discharge is reduced, and the energy utilization rate is increased. The consumption of tap water in the water production process is reduced.
Owner:SPLUNK TECH CONSULTING (BEIJING) CO LTD

Vacuum freeze dryer

The invention discloses a vacuum freeze dryer, particularly relates to the technical field of vacuum freeze drying, and comprises a drying chamber, a refrigerating system, a vacuum system, a heating system, a control system and an energy recovery device. The drying chamber is used for containing materials, the inner wall of the drying chamber is made of efficient heat insulation materials, the refrigerating system comprises a high-precision temperature sensor and an intelligent temperature controller, the refrigerating temperature fluctuation is accurately controlled within + / -1 DEG C, the vacuum degree fluctuation of the vacuum system is kept within + / -5Pa through a vacuum degree sensor and a stabilizing device, and the heating system adopts uniform heating elements. And it is ensured that the heating temperature nonuniformity is smaller than + / -1 DEG C, a control system achieves automatic drying, and an energy recovery device recovers waste cold and waste heat for other links. The problems that an existing vacuum freeze dryer is low in temperature control precision, unstable in vacuum degree, low in automation degree, low in energy utilization rate and the like are solved, the material quality can be effectively kept, the drying efficiency is improved, and energy consumption is reduced.
Owner:SHANGHAI BONUO BIOTECHNOLOGY CO LTD

Edge computing enabled air-to-ground network service deployment and resource allocation method and system

The invention belongs to but is not limited to the technical field of wireless communication, and discloses an edge computing enabled air-ground network service deployment and resource allocation method and system, which is characterized in that application programs are strategically deployed and replaced on a UAV (Unmanned Aerial Vehicle), computing resources and power are reasonably allocated, and on the premise that the task execution efficiency and accuracy are ensured, the resource allocation efficiency is improved. The task execution success rate of the IoT equipment in the intelligent railway system is maximized; the mobile edge computing is introduced into the air-space-ground integrated network, the characteristics of flexible deployment and small transmission delay of the UAV in the air-based network are utilized, and the complexity and dynamism of a traffic system are effectively dealt with and the resource allocation and power control are optimized through collaborative optimization of task placement, cache replacement, a resource allocation strategy and the transmission power of IoT (Internet of Things) equipment. According to the method, a deep reinforcement learning framework is adopted, and an algorithm based on a multi-agent depth deterministic strategy gradient is provided to realize joint optimization of task placement, cache replacement, resource allocation and IoT equipment transmission power.
Owner:XIAN UNIV OF POSTS & TELECOMM

Sludge separating and recycling device

The invention discloses a sludge separating and recycling device, and relates to the technical field of sludge treatment, the sludge separating and recycling device comprises a rack, and a feeding assembly, a separating assembly and a discharging assembly which are mounted above the rack, and the feeding assembly, the separating assembly and the discharging assembly are sequentially arranged from inside to outside. By arranging the feeding assembly, the separating assembly and the discharging assembly which are sequentially installed from inside to outside, sludge can sequentially pass through the feeding assembly, the separating assembly and the discharging assembly in the treatment process, and the feeding assembly and the separating assembly can both drive the sludge to rotate; according to the invention, the centrifugal force is applied to the sludge, so that water, mud and sandstone in the sludge are layered and respectively discharged under the action of the centrifugal force, the efficient separation of the sludge is achieved, the structures which are sequentially arranged from inside to outside can realize the sequential separation of the water, mud and sandstone in the sludge, firstly, the water with a relatively large proportion is separated so as to facilitate the subsequent input of the sludge, and the separation efficiency is improved. Therefore, the sludge treatment efficiency of the device can be improved.
Owner:NANTONG WANGE ENVIRONMENTAL TECH CO LTD

Compression imaging method and system based on Swin Transform network

The invention relates to the technical field of image processing, in particular to a compression imaging method and system based on a Swin Transform network, and the method comprises the steps: carrying out the sampling processing of training data through employing a CC Hadamard matrix, obtaining a coding image, training a neural network based on the Swin Transform through employing the coding image, and obtaining an optimal compression reconstruction model; in the system, the optimal compression reconstruction model is loaded into a processing center, multiple low-resolution images collected by a detector are preprocessed to obtain coded images, and then the optimal compression reconstruction model is utilized to obtain reconstructed images. Compared with a common sampling matrix, the system and the method provided by the invention have the advantages that the reconstruction effect after sampling of the CC Hadamard matrix is better, and the reconstruction quality of the image is effectively improved.
Owner:CHANGCHUN INST OF OPTICS FINE MECHANICS & PHYSICS CHINESE ACAD OF SCI

Correction method and device for helical line deviation of small-modulus plastic cylindrical gear

ActiveCN119130965AComprehensive understanding of geometric featuresHigh geometric accuracyImage enhancementImage analysisHelical lineFeature mining
The invention relates to the technical field of helical line deviation correction, in particular to a correction method and device for helical line deviation of a small-modulus plastic cylindrical gear. The method comprises the following steps: acquiring a multi-azimuth image of a small-modulus gear to be detected; global histogram equalization is carried out on the to-be-detected small-modulus gear multi-azimuth image, and a brightness optimization gear image is constructed; carrying out three-dimensional point cloud reconstruction on the brightness optimized gear image, and carrying out gear total deviation calculation to generate a spiral line total deviation curve; carrying out deviation change situation analysis on the total deviation curve of the spiral line, and carrying out nonlinear shrinkage characteristic mining to obtain a nonlinear shrinkage rule; and performing error compensation calculation on the gear three-dimensional result model according to a nonlinear contraction rule to generate a spiral line contraction error compensation value, and performing error parameter fine adjustment correction on the to-be-detected small-modulus gear so as to obtain corrected gear spiral line parameters. The high-precision and high-accuracy gear helix deviation correction method is realized.
Owner:DONGGUANSHIXINGHUO GEARS CO LTD

Fish feeding intensity evaluation method based on improved VGG16 and spectrogram

The invention provides a fish ingestion intensity evaluation method based on improved VGG16 and a spectrogram, and relates to the field of aquaculture. The method comprises the following steps: acquiring fish ingestion sound signals by adopting a hydrophone, dividing the ingestion sound signals into four types of intensities of'strong ', 'medium', 'weak 'and'no', manually marking the ingestion sound signals, converting the sound signals into spectrograms, and constructing a fish ingestion sound signal data set; an attention mechanism fusion module is designed, the attention mechanism fusion module is integrated on a convolutional neural network, a fish ingestion sound signal data set is used as input, and automatic grading of ingestion intensity is achieved. The method provided by the invention can efficiently evaluate the fish ingestion intensity so as to judge the fish appetite. The method can be applied to fish feeding strategy optimization and behavior research.
Owner:ZHEJIANG UNIV +1

Data interaction processing system for operation panel of high-efficiency 1200nm pulse fiber laser

The invention provides a high-efficiency 1200nm pulse fiber laser operation panel data interaction processing system, and relates to the technical field of data processing, and the system comprises an operation panel module which receives an operation instruction, converts the operation instruction into a digital signal through an instruction processing and conversion model, and transmits the digital signal; the frequency detection module is used for detecting the repetition frequency of the pulse fiber laser to obtain a detection result; the pulse width adjusting module is used for dynamically adjusting the width of the pulse laser according to the information provided by the frequency detection module; and the safety protection module is used for monitoring the working state of the laser, including temperature and current parameters, and carrying out safety early warning. Accurate transmission of the operation instruction, comprehensive monitoring of the laser state, dynamic pulse width adjustment and safety early warning are realized, and the operation convenience, the performance stability and the safety reliability of the laser are improved.
Owner:ZHEJIANG LINGKANG MEDICAL INSTR CO LTD

Interactive three-dimensional medical image segmentation method

The invention relates to a segmentation method for a three-dimensional medical image, in particular to an interactive three-dimensional medical image segmentation method. Comprising the following steps: respectively carrying out preprocessing such as cutting and standardization on to-be-processed data; constructing a neural network based on a visual base model and three-dimensional coding, and in a training stage, inputting training data into the interactive three-dimensional medical image segmentation model for learning; in the test stage, the to-be-segmented data is subjected to online learning through a three-dimensional coding strategy, and further updating of the model is achieved. The invention provides an interactive image segmentation model framework based on a visual base model, an efficient fine tuning strategy and a network structure are adopted, the parameter quantity of the model and the computing power demand of training are reduced, and meanwhile, the high precision and the real-time performance of the model are ensured; a three-dimensional click code extraction strategy is adopted, click information of a user is better input to a network, and the online learning performance of the model is continuously improved.
Owner:UNIV OF CHINESE ACAD OF SCI

Elliptical small roll multi-strip winding equipment and winding process thereof

The invention relates to the technical field of braid winding, and discloses oval small-roll multi-strip winding equipment and a winding process thereof, and the oval small-roll multi-strip winding equipment comprises a table; the two side plates are respectively fixed at two ends of the top surface of the table; the driving structure is arranged on the table; the clamping structure is arranged above the table, the clamping structure is driven by a driving structure, the clamping structure comprises two first clamping plates and two second clamping plates, and the two first clamping plates are arranged right opposite to the two second clamping plates respectively. Through movement of the two first clamping plates and the two second clamping plates and the movable characteristic of the two driving rods, flexible movement of the multiple strip winding pieces is achieved, by means of the design, workers can rapidly replace unused winding plates, tedious disassembly and assembly operation is not needed, and therefore the working efficiency of the whole device is remarkably improved.
Owner:XIAMEN YAMA RIBBONS & BOWS